After flow goes to ai Agent even on 16core cpu spiking 100% and is being proccesed like for 30 minutes, before was about 2-3 minutes?
Anyone has the solution?
Share the output returned by the last node
Information on your n8n setup
n8n version: 1.95.2
Database (default: postgres):
n8n EXECUTIONS_PROCESS setting (default: own, main):
Running n8n via (Docker):
Operating system: Ubuntu 24.04
1 Like
Hey @nikolinm
Can you share yiour workflow please, also, it could be due to a number of issues, have you enabled debug logs / or checked normal logs anything you can see there?
Also, can you try print htop or top from terminal windows plz when this happens, to confirm if it’s node or something else.
You may also try to upgrade to latest version.
Best regards,
Samuel
I have downgraded to 1.94.1 now it’s fine. Maybe was OpenAi Api, I didn’t collect logs.
@nikolinm Ah okay, am sure a newer version this is fixed, but glad it working on that version.
Hope all goes well,
Samuel
Had the same problem and the same solution worked for me.
I also tried higher versions ( 1.95.2-1.97.1) and the problem is the same on all of the versions > 1.94.1
@Edit
Correction - I thought it was fixed - it still is not!
OK, I couldn’t edit the above post, so I am replying to myself. I found a reliable workaround and shared it here:
Alright, I know it’s not ideal to reply to your own message instead of editing, but I thought I’d make an exception (forgive me if this is wrong!) because I think it’s worth it’s own post.
I fixed the issue … kinda … found a good workaround.
I spent a lot of time debugging and I am 100% sure the problem was with (in my case) Basic LLM Chain node. Every single time it was to process too much content, it was freezing the whole docker container.
So, I replaced it with a manual HTTP POST call to …